Constitutive activation of the classical NF-?B pathway, which principally involves RelA/p65, c-Rel, and p50, has been observed in human acute myeloid leukemia (AML) where constitutive NF-?B DNA-binding activity has been detected in myeloid blasts and in CD34? hematopoietic progenitor cells in 47-100% of all cases. In AML, constitutive NF-?B activity could be a direct downstream consequence of activating mutations affecting the PI3 kinase pathway, which is also constitutively active in the majority of AML cases and induces NF-?B through Akt signaling. In mice, constitutive activation of Akt through inducible deletion of the Akt regulatory phosphatase, PTEN, results in rapid hematopoietic stem cell (HSC) loss from bone marrow and development of a transplantable, lethal myeloproliferative disease (MPD) and AML. These studies and others have shown that constitutive activation of Akt and NF-?B distinguishes leukemia-initiating cells (LIC) from normal HSC in bone marrow. To address whether NF-?B activation in HSC is sufficient to induce LIC formation and promote MPD/AML, we expressed a constitutively active allele of I?B kinase beta (Ikk?SS/EE) in normal HSC using a retroviral vector. Analysis of mice reconstituted with Ikk?SS/EE -expressing cells showed that constitutive NF-?B activity promoted relatively rapid HSC loss from bone marrow that, unlike activated Akt, was not accompanied by the development of MPD/AML. NF-?B-mediated HSC loss was likely due to differentiation and loss of HSC self-renewal potential since there was not evidence of increased HSC apoptosis or mobilization to the spleen. Furthermore, blocking NF-?B activation in HSC by conditional deletion of the Ikk regulatory subunit, NEMO, resulted in a substantial decrease in one of the earliest multipotential progenitor subsets (c-Kit?Sca-1? cells) and an increase in absolute numbers of more primitive c-Kit?Sca-1? (KLSF) cells. NEMO loss also completely blocked proliferation of KLSF cells in vitro, which further supports an essential role for NF-?B in promoting the earliest HSC differentiation event. Together, these results demonstrate that constitutive activation of the classical NF-?B pathway, as an initiating event in AML, is not sufficient to maintain HSC or to promote development of LIC associated with AML. The primary goal of this proposal will be to test the central hypothesis that NF-?B functions in normal HSC to couple proliferation and differentiation associated with the first HSC cell division and to maintain viability of LIC subsequent to other oncogenic changes that preserve HSC self-renewal in the presence of activated NF-?B. To test this hypothesis, we will pursue the following specific aims: (1), Determine the function of NF-?B in the regulation of normal HSC maintenance, proliferation, and differentiation (2), Define the contribution of NF-?B to Akt-stimulated leukemogenesis, LT-HSC loss from bone marrow, and to leukemia-initiating cell formation and (3), Characterize whether recurrent cytogenetic abnormalities found in AML, including the translocations AML1-ETO, inv(16), or NUP98-HOXA9, can preserve HSC self-renewal in the presence of activated Akt or NF-?B.